An automatic data mining method to detect abnormal human behaviour using physical activity measurements

Abnormal human behaviour detection under free-living conditions is a reliable method to detect disorders and diseases in healthcare applications. The problem with current methods to detect human behaviour changes is the use of supervised techniques that require human intervention. This work proposes an automatic data mining method based on physical activity measurements. Abnormal human behaviour is detected as an increase or decrease of the physical activity according to the historical data. Human behaviour is evaluated in real time grading its abnormality. The method has been validated involving users with a precision of 100% and a recall of 92%.